inference.sh is a remote MCP server that gives an AI agent 25 tools, among them app_get, app_list, app_run. It answers at https://api.inference.sh/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 10 of 10 times since 4 September 2026; the last check, on 17 October 2026, succeeded.
Run 150+ AI apps — image, video, audio, LLMs, 3D and more. Browse, execute, stream results.
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
app_get | Get detailed app info including input/output schemas and version history Inputs: app |
app_list | Search and filter available apps on inference.sh Inputs: category, cursor, limit, query |
app_run | Run an app on inference.sh. Creates a task that executes the app with the given input. Inputs: app, function, infrastructure, input |
knowledge_create | Create a new knowledge entry (concept, observation, preference, reference, etc.) Inputs: content, description, name, type |
knowledge_delete | Delete a knowledge entry by ID Inputs: id |
knowledge_get | Get detailed information about a knowledge entry by namespace/name or ID Inputs: ref |
knowledge_list | List your knowledge entries, optionally filtered by type Inputs: cursor, limit, type |
knowledge_search | Search your knowledge entries by query Inputs: cursor, limit, query, type |
mcp_connect | Initiate a connection to an MCP server. For OAuth servers, returns an authorization URL to open in a browser. Inputs: slug |
mcp_disconnect | Remove connection to an MCP server Inputs: slug |
mcp_get | Get details about an MCP server including connection status Inputs: slug |
mcp_list | Browse available MCP servers that can be connected to Inputs: category, cursor, featured, limit |
mcp_run | Call a tool on a connected MCP server. The server must be connected first via mcp_connect. Inputs: input, slug |
mcp_search | Search available MCP servers by query Inputs: cursor, limit, query |
mcp_tools | List available tools on a connected MCP server, or get details for a specific tool Inputs: slug |
skill_files | List supplementary files for a skill (references, scripts, etc.) Inputs: ref |
skill_search | Search the public skill store Inputs: category, cursor, limit, query |
skill_store | Browse approved skills in the public skill store Inputs: category, cursor, featured, limit |
skill_upload | Create or update a skill with SKILL.md content and optional supporting files Inputs: allowed_tools, content, description, files, name |
skill_use | Fetch a skill's assembled SKILL.md content. Tries the inference.sh store first, falls back to GitHub. Returns the full skill content for the agent to consume. Inputs: ref |
skill_view | View a specific supplementary file from a skill's manifest Inputs: path, ref |
task_cancel | Cancel a running or queued task Inputs: force, task_id |
task_get | Get status and output of a task Inputs: task_id |
task_list | List your tasks on inference.sh Inputs: cursor, limit, status |
task_logs | Get execution logs for a task Inputs: log_type, task_id |
Every block below points at https://api.inference.sh/mcp. Pick your client and copy.
claude mcp add --transport http inference-sh https://api.inference.sh/mcp
{
"mcpServers": {
"inference-sh": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://api.inference.sh/mcp"]
}
}
}{
"mcpServers": {
"inference-sh": {
"url": "https://api.inference.sh/mcp"
}
}
}{
"servers": {
"inference-sh": {
"type": "http",
"url": "https://api.inference.sh/mcp"
}
}
}{
"context_servers": {
"inference-sh": {
"url": "https://api.inference.sh/mcp"
}
}
}{
"mcpServers": {
"inference-sh": {
"serverUrl": "https://api.inference.sh/mcp"
}
}
}{
"mcpServers": {
"inference-sh": {
"url": "https://api.inference.sh/mcp"
}
}
}name: inference.sh
version: 0.0.1
schema: v1
mcpServers:
- name: inference.sh
type: streamable-http
url: https://api.inference.sh/mcp{
"mcpServers": {
"inference-sh": {
"httpUrl": "https://api.inference.sh/mcp"
}
}
}{
"mcpServers": {
"inference-sh": {
"type": "http",
"url": "https://api.inference.sh/mcp"
}
}
}{
"mcpServers": {
"inference-sh": {
"url": "https://api.inference.sh/mcp"
}
}
}goose://extension?url=https%3A%2F%2Fapi.inference.sh%2Fmcp&type=streamable_http&id=inference-sh&name=inference.sh
It answered our last check on 17 October 2026 in 321 ms. Over 10 checks since 4 September 2026 it answered 10. Status here comes from Neuronto's own probes, not from the publisher.
Yes. It is published by inference.sh and served from api.inference.sh, the vendor's own domain: https://api.inference.sh/mcp.
25, as returned by the server's own tools/list on 7 September 2026: app_get, app_list, app_run, knowledge_create, knowledge_delete, knowledge_get, knowledge_list, knowledge_search, and 17 more.
It answered our last check on 17 October 2026 in 321 ms. Over 10 checks since 4 September 2026 it answered 10.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://api.inference.sh/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.
It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.
Publisher: ac.inference.sh. Endpoint host: api.inference.sh. ARD identifier: urn:air:ac.inference.sh:mcp:mcp.
First indexed 31 August 2026. The tools are exactly what the server returned; nothing on this page
is inferred from its name. Ranking in Neuronto search is never for sale.